Epidemiological and experimental studies strongly support the association between smoking and cancer of cervix uteri. However, until now, genotoxic, tobacco-related compounds have not been detected in this tissue. The goal of this proposal, which combines laboratory studies with clinical investigations, is to unequivocally identify and quantify tobacco-related carcinogens in the human cervical mucus and characterize and quantify DNA adducts resulting from the exposure of cervical epithelium to these carcinogens or their metabolites. The hypothesis being tested is that tobacco-specific nitrosamines and polynuclear aromatic hydrocarbons (PAH) may be involved in the etiology of cervical cancer. The specific aims are: (1) Quantify the tobacco- specific nitrosamines (TSNA), especially 4-(methylnitrosamino)-1 -(3- pyridyl)-1-butanone (NNK) and its major metabolite 4- (methylnitrosamino)-1-(3-pyridyl)-1-butanol (NNAL) in cervical mucus of healthy, active smokers and of women possibly exposed to environmental tobacco smoke (ETS) and test the hypothesis of endogenous formation of TSNA from nicotine in human cervix. (2) Determine the capacity of human cervical epithelium to metabolize TSNA, leading to the formation of DNA adducts, and assess the levels of these adducts in human cervical epithelium (3) Characterize and quantify levels of PAH diol epoxide-derived DNA adducts by a gas chromatography-mass spectroscopy (CC-MS) in surgical specimens. (4) Quantify environmental B[a]P, fluoranthene, chrysene, and benzo[a]anthracene (B[a]A), and their metabolites in cervical mucus of smokers vs nonsmokers by supercritical fluid extraction SFE and CC-MS. The results of these investigations will provide new insights on the association between smoking and cervical cancer as well as new biomarkers for future epidemiological studies.